Search Results (330)

Microplastics (MPs) are widely recognized as emerging contaminants in freshwater environments. Their identification often relies on extensive sample preparation and chemical treatments, which increase analysis time, reagent use, and overall resource consumption. Consequently, there is a growing need for sustainable analytical approaches enabling reliable MP detection while minimizing sample handling. This study proposes an analytical workflow based on hyperspectral imaging (HSI) as a proof-of-concept approach for direct identification of MPs in freshwater samples. Water samples collected from three different rivers, containing heterogeneous natural materials, were spiked with MPs (250–1000 μm) of three common polymers, namely high-density polyethylene (HDPE), polystyrene (PS), and polypropylene (PP), to simulate realistic contamination scenarios. HSI acquisitions were performed in the short-wave infrared range (SWIR: 1000–2500 nm). Spectral preprocessing and principal component analysis (PCA) were applied for data exploration, while a hierarchical partial least squares-discriminant analysis (Hi-PLS-DA) model was developed to classify five target classes: natural materials, water, HDPE, PS, and PP. Despite sample complexity, the proposed workflow achieved satisfactory classification results, as demonstrated by the predicted class map and the corresponding statistical metrics (sensitivity, specificity, precision, and F1-score: 0.900–0.999). These results highlight the potential of the SWIR-HSI-based approach as a rapid and sustainable method for direct MP identification in freshwater samples and provide methodological insights for rapid MP screening strategies requiring minimal sample preparation. Full article

Accurate and automated extraction of solid waste sites from remote-sensing imagery constitutes a pivotal demand for contemporary environmental regulation and risk mitigation. However, in high-resolution remote-sensing imagery, solid waste sites are typically represented as a single semantic image object (SIO), which is composed of multiple physical image parcels (PIPs) exhibiting significant variations in scale, morphology, and spectral properties. This intrinsic heterogeneity substantially increases the complexity and uncertainty of multi-class site identification. To address this challenge, this paper proposes WasteMOE Net (WMN), which is developed based on the core concept of modeling the SIO–PIP relationship. WMN adopts a heterogeneous expert selection mechanism combined with a nested mixture-of-experts architecture. It thus enables adaptive perception of complex PIPs across diverse scenarios and their integrated discrimination at the SIO level. In addition, by incorporating the explicit nonlinear representation capability of the KAN network, WMN effectively improves multi-class recognition accuracy while maintaining computational efficiency. Furthermore, this study constructs a high-resolution solid waste site dataset in accordance with the SIO–PIP-aware annotation principle, encompassing five representative categories: tailings ponds (TP), construction spoil sites (CSS), landfill sites (LS), garbage dump sites (GDS), and excavation sites (ES). Experimental results show that WMN achieves mAP50 values of 74.2% (GDS), 63.5% (CSS), 80.9% (ES), 85.4% (TP), and 83.1% (LS) in detection tasks, and 75.4%, 64.1%, 83.0%, 86.7%, and 84.1% for the corresponding categories in segmentation tasks. It achieves competitive performance compared with state-of-the-art methods in both tasks. Further, in a real-world application over Loudi City, China, WMN completed the processing of a 490.67 km² area within 1.34 h. The recognition accuracies for GDS and ES reached 54.8% and 65.3%, respectively. Finally, the proposed method has been successfully integrated into a GIS-based solid waste pollution risk prevention system, which markedly boosts the overall efficiency of environmental monitoring and on-site inspections. Full article

Wheat is a major staple crop, and storage mold growth poses a severe threat to grain safety and quality stability. Natural mold development in stored wheat exhibits subtle, localized, and highly heterogeneous characteristics. Existing unimodal methods and global fusion approaches generally suffer from insufficient local feature sensitivity, hindering fine-grained mold severity grading. To address this limitation, we propose a Mask-Guided Fine-Grained Fusion Network, a weakly supervised framework based on local RGB–HSI fusion. This framework employs a dynamic parallel A/B experimental design to construct time-matched proxy labels via weakly supervised learning. A standardized preprocessing pipeline including single-kernel extraction, foreground segmentation, and cross-modal registration is established to resolve RGB–HSI spatial misalignment, ensuring physical-level spatial consistency of multimodal features. The model incorporates a Foreground-Aware Spectral Recalibration (FASR) module to suppress background noise, a Mask-Guided Dilated Cross-modal Local Attention (MDCLA) mechanism to establish fine-grained local mappings between RGB visual phenotypes and hyperspectral responses, and a sample-level adaptive fusion strategy to dynamically weight features by modal reliability, enhancing representation of complex samples across all mold stages. Experiments show that the Mask-Guided Fine-Grained Fusion Network achieves 0.9689 classification accuracy, 0.9698 Macro-F1 score, and 0.0593 Mean Absolute Error (MAE), significantly outperforming state-of-the-art unimodal deep models and global attention fusion baselines. This work provides a proof-of-principle framework for fine-grained non-destructive mold risk assessment in stored wheat. Full article

Hydrothermal alteration mapping is a critical component of porphyry copper exploration because alteration assemblages provide important vectors toward mineralization. This study presents a systematic evaluation of supervised machine learning algorithms for delineating hydrothermal alteration zones using Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) short-wave infrared (SWIR) surface reflectance data (AST_07XT). The investigation focuses on the Nain region within the central Urumieh–Dokhtar Magmatic Arc (UDMA), Iran, a major metallogenic belt hosting numerous porphyry copper systems. Representative spectral endmembers corresponding to Al–OH-bearing and Mg–OH-bearing hydrothermal alteration minerals were extracted using Minimum Noise Fraction (MNF), Pixel Purity Index (PPI), and n-dimensional visualization techniques. These endmembers were subsequently used to train and evaluate a comprehensive suite of supervised machine learning classifiers, including linear, kernel-based, tree-based, ensemble, probabilistic, boosting, and neural-network algorithms for pixel-wise hydrothermal alteration mapping. Model performance was evaluated using multiple statistical metrics, including overall accuracy (OA), average accuracy (AA), precision, recall, F1-score, Cohen’s kappa coefficient, area under the ROC curve (AUC), spatial cross-validation accuracy, uncertainty analysis, and spatial agreement analysis. Among the evaluated classifiers, SVM_Linear, SVM_RBF, LDA, and MLP achieved the highest classification performance, with overall accuracies exceeding 94% and strong spatial consistency between classified maps. The resulting alteration maps display spatially coherent distributions of Al–OH and Mg–OH minerals that are consistent with established hydrothermal alteration zoning models in porphyry–epithermal systems. The mapped hydrothermal alteration zones show strong spatial correspondence with known mineralized areas and alteration patterns within the Urumieh–Dokhtar Magmatic Arc, confirming the geological reliability of the classification results. Uncertainty analysis further indicates high model confidence across most alteration zones, with higher uncertainty values mainly restricted to transitional and spectrally heterogeneous regions. The results demonstrate that integrating ASTER SWIR imagery with supervised machine learning algorithms provides a robust, scalable, and transferable framework for regional-scale hydrothermal alteration mapping and mineral exploration in porphyry copper provinces. Full article

Tropical rainforests are characterized by high species diversity and complex canopy structure, making accurate forest stand classification important for ecosystem assessment, biodiversity monitoring, and forest carbon estimation. However, single-source remote sensing data lacks sufficient discrimination ability to address the issue of spectral similarity among classes, and conventional convolutional neural networks often struggle to extract discriminative features and integrate heterogeneous data in highly complex forests. To address these challenges, this study developed a dual-branch deep learning framework that integrates DenseNet and ConvNeXt for classification in Hainan Tropical Rainforest National Park. The framework combines sub-meter Google Earth imagery to capture spatial–textural detail with multi-temporal Sentinel-2 imagery to represent phenological variation. The results showed that multi-temporal Sentinel-2 data outperformed single-date imagery by capturing phenological patterns, and that the fusion of high-resolution spatial information and multi-temporal spectral information yielded higher accuracy than either data source alone. The dual-branch model achieved an overall accuracy of 94.47% and a Kappa coefficient of 0.94, outperforming all benchmark models. These findings indicate that branch-specific feature extraction and adaptive fusion can improve fine-scale classification in complex tropical rainforest environments. The proposed framework provides a practical approach for fine-scale forest stand mapping and may support biodiversity monitoring, ecological assessment, and sustainable forest management. Full article

Rapid identification of maize waterlogging is essential for post-disaster agricultural assessment, but most existing methods rely on multi-temporal imagery that is often unavailable immediately after extreme rainfall events. This study proposes SAB-DeepLabV3+, a semantic segmentation model for mapping waterlogging-affected maize from single-date multispectral imagery within pre-extracted maize planting areas. Built on DeepLabV3+, the model integrates three task-specific modules: a Spectral-Spatial Information Enhancement Module to improve feature discrimination under spectral mixing, an Adaptive Multi-Scale Pooling Module to capture heterogeneous patch sizes, and a Boundary Enhancement Module to refine transition zones. A pixel-level dataset containing 12,198 image patches was constructed from 62 multispectral scenes collected across five major maize-producing cities in Heilongjiang Province, China, during 2022–2024. On the test set, SAB-DeepLabV3+ achieved a waterlogged-class IoU of 68.30%, mIoU of 80.37%, mF1 of 88.62%, and OA of 93.49%, outperforming DeepLabV3+. Leave-one-city-out evaluation further produced an average mIoU of 76.56% and a waterlogged-class IoU of 63.45%. These results indicate that single-date high-resolution multispectral imagery can support rapid and reliable maize waterlogging mapping. Full article

Forests of Lualaba Province (DR Congo) form a compositionally complex mosaic of dry dense forest, gallery forest, and Miombo woodland. Yet, categorical land-cover maps impose discrete boundaries on these inherently continuous vegetation gradients, systematically discarding subpixel compositional information critical for forest monitoring and carbon accounting. The magnitude of this information loss at the landscape scale, however, remains largely unquantified. In this study, we train a Multi-Output Neural Network (MONN) using Sentinel-2 spectral and textural predictors (2025) to estimate the proportional cover of three forest types across the province. Model performance is benchmarked against a normalised Random Forest (RF) using spatial block cross-validation. Categorical information loss is quantified pixel-wise using two complementary metrics, dominant class proportion and Shannon compositional entropy, alongside a derived interpretive quantity, categorical information loss. The MONN slightly outperformed RF (R² = 0.648 vs. 0.630; RMSE = 0.224 vs. 0.229), yet the results reveal a fundamentally heterogeneous landscape structure. The mean dominant-class proportion was only 56.2%, indicating that categorical maps discard, on average, 43.8% of compositional information per pixel. Only 7.9% of forested pixels exceeded the 75% dominance threshold, while Shannon entropy reached 74.1% of its theoretical maximum, indicating that forest types coexist in near-equal proportions across most pixels. This renders categorical attribution structurally inadequate for most of the forested landscape. Across 92.1% of forested pixels, no single forest type achieved clear dominance. These results show that compositional mixing is the dominant structural condition of the landscape, and that compositional mapping is essential for representing tropical forest structure in heterogeneous drylands. By formally quantifying categorical information loss at the landscape scale, this study shows that continuous compositional mapping converts this structural ambiguity into a spatially explicit ecological signal, with direct implications for monitoring vegetation dynamics and biodiversity, suggesting a structural source of error in carbon stock estimation in tropical dry forests that warrants empirical validation. Full article

Satellite-based vegetation monitoring has evolved from mapping vegetation canopy properties at single points in time toward analysing time-resolved dynamics of vegetation traits and process-related variables. Retrieved traits and solar-induced chlorophyll fluorescence (SIF) are inherently defined by sensor-specific spatial resolution, temporal integration, and spectral response. Modifying these characteristics alters the retrieval problem itself: under nonlinear retrievals and heterogeneous landscapes, aggregation and retrieval are generally non-commutative, and error components scale differently with resolution. Consequently, increasing spatial, spectral, or temporal detail does not guarantee improved ecological accuracy; a phenomenon we term the resolution–accuracy paradox. These interacting processes define the effective scale of vegetation products, which may differ from nominal sensor resolution and governs the interpretation of retrieved vegetation traits. When products with differing resolutions or compositing strategies are combined, scale effects can induce systematic artefacts in spatial patterns and derived dynamic indicators that cannot be resolved through improved per-pixel accuracy alone. This review establishes a scale-aware conceptual framework that treats scale as an explicit diagnostic dimension linking observation characteristics, retrieval formulations, trait definitions, and aggregation operators. We analyse how scale interactions influence spatial patterns, temporal dynamics, disturbance signals, and multiresolution data fusion, and derive diagnostic principles, best-practice guidelines, and research priorities for the scale-consistent interpretation of vegetation trait dynamics and SIF-constrained productivity and stress indicators across spatial and temporal scales. Framed in the context of upcoming hyperspectral missions such as CHIME and FLEX, which increase spectral information content, robust interpretation of vegetation products requires scale-consistent analysis and uncertainty-aware processing. For practitioners, this implies that vegetation products should be interpreted, validated, and compared at their effective scale rather than assuming that a finer spatial, spectral, or temporal resolution necessarily yields more reliable ecological information. Full article

Phosphate waste rock piles (PWRPs) generated by open-pit phosphate mining are highly heterogeneous and difficult to characterize using conventional point sampling alone, which limits representative resource assessment, selective recovery, and rehabilitation planning. This study develops an integrated framework combining PRISMA spaceborne hyperspectral imagery, ground-based mineralogical analyses, and spatially constrained machine learning to map lithological heterogeneity at the Benguerir phosphate mining site, Morocco. A three-stage spectral optimization workflow, including atmospheric band masking, Savitzky–Golay filtering, and analysis of variance (ANOVA)-based feature selection, was applied to identify the most discriminative Short-Wave Infrared (SWIR) bands for lithological classification. After removing redundant observations located within shared PRISMA pixel footprints, 127 spatially independent samples were retained for model development. Five supervised classifiers (Random Forest, Extra Trees, XGBoost, Support Vector Machine, and K-Nearest Neighbors) were evaluated under a spatially constrained cross-validation framework aligned with the 30 m native PRISMA pixel size. Ensemble-based models, especially Extra Trees and Random Forest, provided the most stable performance, with balanced accuracies of 0.56–0.69 and area under the receiver operating characteristic curve (AUC) values exceeding 0.95 for carbonate-dominated lithologies. Lower discrimination between phosphate and siliceous facies reflects intrinsic mineralogical mixing and spectral overlap at the sensor scale. Entropy-based uncertainty and posterior probability mapping revealed spatially structured prediction ambiguity concentrated along lithological boundaries and transitional zones, consistent with petrographic evidence of compositional heterogeneity. These results indicate that moderate but stable accuracies likely represent realistic performance limits for spaceborne hyperspectral mapping of complex mining environments under spatial constraints. The proposed framework provides a transferable and uncertainty-aware basis for lithological mapping, selective recovery assessment, and sustainable phosphate waste management. Full article

Vegetation phenology refers to the cyclical growth patterns of vegetation in nature, which are influenced by climatic conditions, human activities, and genetic factors. It plays an irreplaceable role in regulating carbon cycling and energy flow within natural ecosystems. However, the combination of a cloudy and rainy climate with a landscape characterized by the interplay of land and water and fragmented patches has long posed challenges for remote sensing phenological monitoring data, including a scarcity of valid observations, frequent temporal gaps, and spectral distortion in mixed pixels. These issues make it difficult to reliably support the needs of wetland phenological inversion and mapping. To address this issue, this study uses vegetation inversion in the Poyang Lake wetlands as a case study and reconstructs high-spatiotemporal-resolution time-series kNDVI data based on multi-source remote sensing data. Methodologically, we propose an improved and enhanced spatiotemporal adaptive reflectance fusion model, IESTARFM. This model enhances the homogeneity of similar pixel selection through adaptive matching windows and land cover constraints. Additionally, it explicitly incorporates cloud probability and time-lag factors into the weighting structure to systematically downweight unreliable observations, and further employs quadratic term corrections to account for the nonlinear growth response of kNDVI. Using the reconstructed dataset, key phenological information is extracted by combining third-order harmonic analysis with a dynamic thresholding method, thereby enhancing the robust characterization of seasonal trajectories under conditions of missing data and noise. Accuracy evaluation results show that the 10m/8d high-frequency kNDVI dataset reconstructed by IESTARFM achieves at least a 12.61% improvement in fusion accuracy compared to classical methods such as ESTARFM, STARFM, and FSDAF, with a maximum reduction in RMSE of 0.026, and effectively restores details in areas with thin cloud cover. The reconstructed kNDVI series achieved a coefficient of determination $R^2$ = 0.875 and RMSE = 0.066 relative to Sentinel-2 observations, indicating that the reconstructed series closely reproduces the reference imagery in both amplitude and spatial structure. The phenological parameters derived from kNDVI exhibit an RMSE of 4.81 days compared to field observations, demonstrating that the reconstructed time series reliably captures the timing of key phenological events. It should be noted that the proposed approach is designed for post-event time-series reconstruction and is not intended for real-time forecasting. In summary, this study collaboratively enhanced the reliability of high-resolution index time-series reconstruction and phenological identification in cloudy and rainy wetlands through three key aspects: cloud noise suppression, heterogeneous boundary preservation, and nonlinear growth characterization. It provides a generalizable technical foundation for dynamic monitoring of wetland vegetation, ecological restoration assessment, and refined management in regions with frequent cloud and rainfall. Full article

Accurate mapping of offshore aquaculture remains challenging in complex coastal environments due to heterogeneous backgrounds, variable sea states, blurred pond boundaries, adhesion among densely distributed cages, and the weak texture of floating rafts. To address these limitations, this study proposes a bi-temporal multimodal cross-fusion network (BMCF-Net) for fine-scale offshore aquaculture segmentation from Sentinel-1/2 imagery. The framework jointly exploits bi-temporal observations acquired during non-ice and sea-ice periods and integrates them through a bi-temporal fusion module to enhance target–background separability and suppress environmental noise. In addition, a feature correction module and a multi-head feature fusion module are introduced to strengthen cross-modal alignment between SAR structural information and optical spectral–textural cues, thereby improving the separation of dense aquaculture units and the detection of weak-texture targets. Experiments conducted on a multimodal dataset from the Liaoning coastal zone show that BMCF-Net achieves F1-scores of 93.15%, 93.90%, and 89.04% for aquaculture ponds, cages, and floating rafts, respectively, outperforming state-of-the-art segmentation models such as FTransUNet. The proposed model was further applied to produce a high-resolution aquaculture distribution map for Liaoning Province in 2023 and to analyze area dynamics over the past six years. The results demonstrate the potential of BMCF-Net for large-scale offshore aquaculture monitoring and coastal resource management. Full article

The spatial heterogeneity and dynamic changes in soil organic matter (SOM) are key indicators for assessing cultivated land quality and the carbon cycle. Currently, large-scale SOM monitoring relies primarily on limited ground sampling, making it difficult to capture continuous spatiotemporal variation patterns. Taking Anhui Province, China as the study area, this research integrates multi-source remote sensing and geostatistical methods to construct a multi-source collaborative SOM inversion model and analyze its spatiotemporal evolution patterns, thereby achieving high-precision, continuous spatiotemporal monitoring of SOM. A total of 3026 sampling points in Huangshan, Chuzhou and Fuyang cities in Anhui Province were selected as model training samples. The study divided the terrain into three elevation zones (<20 m, 20–40 m, >40 m) and employed the Synthetic Minority Oversampling Technique (SMOTE) method to optimize sample distribution. Based on MODIS data, this study screened spectral bands and key phenological periods significantly correlated with SOM. By integrating spectral information from Landsat 8/9 OLI imagery, meteorological data and topographic factors, a random forest (RF) inversion model incorporating multi-source environmental variables was constructed. The results indicate that (1) the RF-based SOM inversion model exhibits moderate predictive accuracy acceptable for regional-scale SOM mapping, with a coefficient of determination (R²) of 0.55 and a root-mean-square error (RMSE) of 3.3 g/kg, effectively enabling the quantitative estimation of SOM at a regional scale. (2) The model’s inversion results reflect the spatial distribution of SOM in cultivated land in Anhui Province for the years 2019, 2022 and 2024. The provincial average SOM value shows an upward trend, with SOM content exhibiting a pattern of higher levels in the south and lower levels in the north, higher levels in the west and lower levels in the east, as well as a tendency to cluster. (3) Analysis using GeoDetector indicates that topography and precipitation are the primary drivers influencing SOM distribution, and the interaction between these two factors provides significantly greater explanatory power for SOM distribution than either factor alone. Through the integration of multi-source remote sensing data and model optimization, this study has validated the feasibility of multi-scale remote sensing-based SOM inversion, revealed the spatial differentiation characteristics and driving mechanisms of SOM in Anhui Province’s cultivated land, and provided a scientific basis for improving cultivated land quality and soil carbon sink management. Full article

This study examines long-term mangrove vegetation dynamics in the Cananéia–Iguape Coastal System (CICS), southeastern Brazil, with emphasis on their relevance for coastal ecosystem monitoring and sustainability. Land-use and land-cover (LULC) data from MapBiomas were combined with MODIS-derived vegetation indices, namely the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI), and Fractional Vegetation Cover (FVC) to assess spatial variability and temporal trends from 2003 to 2024. Spatial anomalies were calculated as deviations from long-term mean conditions, whereas temporal trajectories were evaluated using the non-parametric Mann–Kendall test and Sen’s slope estimator. The results indicate limited spatial variability, with 98.35% of the study area for the NDVI and 99.51% for the EVI showing no detectable deviations from long-term averages. Within mangrove areas, however, statistically significant positive trends were identified for the NDVI (Z_MK = 2.43; p = 0.02), EVI (Z_MK = 2.03; p = 0.04), and FVC (Z_MK = 2.43; p = 0.02), suggesting a gradual increase in spectral greenness and FVC-derived vegetation density. The moderate correlation between mangrove extent and the NDVI (r = 0.61; p < 0.05) indicates that the mapped mangrove area is partially associated with variations in spectral greenness, although this relationship should not be interpreted as direct evidence of ecological recovery or improved ecosystem conditions. Overall, the findings point to low-magnitude but consistent vegetation index changes in a predominantly stable mangrove system. The integration of LULC information, spectral indices, and FVC provides a consistent regional-scale basis for interpreting mangrove dynamics in heterogeneous coastal environments and for guiding long-term monitoring efforts. Full article

Color measurement and monitoring of chromatic changes over time play a key role in the study and conservation of historical materials. In this context, hyper-spectral imaging (HSI) offers spatially resolved spectral information that can be converted into colorimetric data, although its quantitative reliability under in situ conditions remains challenging. This study evaluates the colorimetric performance of a HSI system (Specim IQ) through comparison with a reference spectrocolorimeter (Konica-Minolta CM-700d), combining laboratory measurements on certified standards and in situ analyses on purple-dyed manuscripts. Colorimetric coordinates (CIELAB) and color differences (ΔE₀₀) were used to assess accuracy, precision, and systematic deviations. Under controlled conditions, HSI showed good agreement with reference measurements, although systematic biases were observed. In situ applications revealed reduced accuracy (average ΔE₀₀ ≈ 4.3) due to material heterogeneity and acquisition constraints. Despite these limitations, HSI preserved consistent relative chromatic relationships, enabling meaningful comparative analysis. Spatially resolved mapping of colorimetric parameters proved effective for visualizing chromatic variability, dye distribution, and degradation patterns. These results demonstrate that, while not fully reliable for absolute colorimetric assessment in situ, HSI represents a powerful tool for non-invasive, spatially resolved color analysis of complex historical materials. Full article

Accurate mapping of winter rapeseed cultivation areas is crucial for food security assessment and agricultural resource management, yet remains a persistent challenge in mountainous regions characterized by complex topography and highly fragmented field parcels. To address these challenges, this study develops a progressive deep learning framework using single growing-season data from the Hanzhong Basin. We conducted a structured comparison of remote sensing indices, machine learning, and deep learning approaches for rapeseed identification in heterogeneous landscapes. First, sensitivity analysis of the Flowering Index for Rapeseed was performed to identify the optimal parameterization, yielding high inter-class separability (ND = 0.959) during peak flowering and a threshold-based overall accuracy (OA) of 94.41%. Second, a multidimensional feature space was constructed by integrating Sentinel-2 spectral bands, image texture metrics, and topographic variables; Random Forest-based feature importance selection subsequently enhanced Support Vector Machine classification performance to an OA of 90.70%. Third, we proposed an innovative three-stage progressive UNet++ architecture: Stage1 focuses on binary rapeseed/non-rapeseed classification to establish spatial priors; Stage2 refines discrimination among spectrally similar vegetation classes (rapeseed and other vegetation); and Stage3 achieves comprehensive seven-class semantic segmentation. A weighted focal loss function combined with a weight inheritance mechanism was employed to mitigate class imbalance and facilitate inter-stage knowledge transfer. The final model attained an OA of 98.65% and a mean intersection over union of 95.29%, while effectively suppressing salt-and-pepper noise artifacts in geometrically fragmented parcels. Our findings demonstrate the substantial advantages of progressive deep learning strategies for crop monitoring in topographically constrained environments. Full article